Embedding Utils

Utilities for generating text embeddings.

logger module-attribute

logger = getLogger(__name__)

generate_embedding

generate_embedding(
	texts: list[str], config_loader: ConfigLoader
) -> list[list[float]]

Generate embeddings for a list of texts using model2vec.

Parameters:

Name	Type	Description	Default
texts	list[str]	List of text strings to embed.	required
config_loader	ConfigLoader	ConfigLoader instance used to load embedding model configuration.	required

Returns:

Type	Description
list[list[float]]	List of embeddings (each embedding is a list of floats)

Source code in src/codemap/processor/utils/embedding_utils.py

def generate_embedding(texts: list[str], config_loader: "ConfigLoader") -> list[list[float]]:
	"""
	Generate embeddings for a list of texts using model2vec.

	Args:
		texts: List of text strings to embed.
		config_loader: ConfigLoader instance used to load embedding model configuration.

	Returns:
		List of embeddings (each embedding is a list of floats)

	"""
	with progress_indicator("Loading model..."):
		from model2vec import StaticModel

		model_name = config_loader.get.embedding.model_name
		model = StaticModel.from_pretrained(model_name)

	with progress_indicator("Generating embeddings..."):
		try:
			embeddings = model.encode(texts)
			return embeddings.tolist()  # Convert np.ndarray to list of lists
		except Exception:
			logger.exception("Error generating embeddings")
			raise